Skip to content

tonickkozlov/twitter-company-rating-backend

Repository files navigation

Goal

This project aims to aggregate company customer satisfaction score based on mentions of company's account on Twitter and textual analysis.

Current working instance aggregates rating for some companies from the Fortune 500 list, and can be found here.

Implementation

There are 3 components to the system, implemented as independent microservices with docker:

All services and code is meant to be run from the root of the project, see details below.

Testing

python3 -m pytest */tests/test_*.py

Initialization

Before the Synchronization service can be run continuously, it has to be initialized with a list of companies:

docker-compose run sync_service python3 -m sync_service init companies.full.txt

This command will go through company names listed in companies.full.txt, will try to find company's Twitter account by name, and add it to the DB

Running

docker-compose up

Will start the API server at :5000 as well as begin fetching tweets for companies from Twitter.

Releases

No releases published

Packages

No packages published

Languages